qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-11-22 05:34:39 -07:00
Code Releases Activity

strconv: implement fast path for rounding already short numbers.

Browse Source 
benchmark                   old ns/op   new ns/op   delta
BenchmarkFormatFloatDecimal      3765        1386    -63%

R=rsc
CC=golang-dev, remy
https://golang.org/cl/5494060
This commit is contained in:
Rémy Oudompheng 2012-01-12 11:34:06 -08:00 committed by Russ Cox
parent f0f6aa59cc
commit 37cd165838
1 changed files with 16 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
src/pkg/strconv/ftoa.go
View File

@ -178,15 +178,26 @@ func roundShortest(d *decimal, mant uint64, exp int, flt *floatInfo) {
return return
} }
// TODO(rsc): Unless exp == minexp, if the number of digits in d
// is less than 17, it seems likely that it would be
// the shortest possible number already. So maybe we can
// bail out without doing the extra multiprecision math here.
// Compute upper and lower such that any decimal number // Compute upper and lower such that any decimal number
// between upper and lower (possibly inclusive) // between upper and lower (possibly inclusive)
// will round to the original floating point number. // will round to the original floating point number.
// We may see at once that the number is already shortest.
//
// Suppose d is not denormal, so that 2^exp <= d < 10^dp.
// The closest shorter number is at least 10^(dp-nd) away.
// The lower/upper bounds computed below are at distance
// at most 2^(exp-mantbits).
//
// So the number is already shortest if 10^(dp-nd) > 2^(exp-mantbits),
// or equivalently log2(10)*(dp-nd) > exp-mantbits.
// It is true if 332/100*(dp-nd) >= exp-mantbits (log2(10) > 3.32).
minexp := flt.bias + 1 // minimum possible exponent
if exp > minexp && 332*(d.dp-d.nd) >= 100*(exp-int(flt.mantbits)) {
// The number is already shortest.
return
}
// d = mant << (exp - mantbits) // d = mant << (exp - mantbits)
// Next highest floating point number is mant+1 << exp-mantbits. // Next highest floating point number is mant+1 << exp-mantbits.
// Our upper bound is halfway inbetween, mant*2+1 << exp-mantbits-1. // Our upper bound is halfway inbetween, mant*2+1 << exp-mantbits-1.
@ -200,7 +211,6 @@ func roundShortest(d *decimal, mant uint64, exp int, flt *floatInfo) {
// in which case the next lowest is mant*2-1 << exp-mantbits-1. // in which case the next lowest is mant*2-1 << exp-mantbits-1.
// Either way, call it mantlo << explo-mantbits. // Either way, call it mantlo << explo-mantbits.
// Our lower bound is halfway inbetween, mantlo*2+1 << explo-mantbits-1. // Our lower bound is halfway inbetween, mantlo*2+1 << explo-mantbits-1.
minexp := flt.bias + 1 // minimum possible exponent
var mantlo uint64 var mantlo uint64
var explo int var explo int
if mant > 1<<flt.mantbits || exp == minexp { if mant > 1<<flt.mantbits || exp == minexp {